A number of vision-based methods for detecting laser-induced defects on optical components have been implemented to replace the time-consuming manual inspection. While deep-learning-based methods have achieved state-of-the-art performances in many visual recognition tasks, their success often hinges on the availability of a large number of labeled training sets. In this paper, we propose a surface defect detection method based on image segmentation with a U-shaped convolutional network (U-Net). The designed network was trained on paired sets of online and offline images of optics from a large laser facility. We show in our experimental evaluation that our approach can accurately locate laser-induced defects on the optics in real time. The main advantage of the proposed method is that the network can be trained end to end on small samples, without the requirement for manual labeling or manual feature extraction. The approach can be applied to the daily inspection and maintenance of optical components in large laser facilities.
We report a kind of structured light beam with periodical polarization and phase singularities. It is generated from a setup consisting of conical refraction transformation and 4f-system. By this setup, the periodical structures are produced without any change of intensity distributions. We analyze both theoretically and experimentally the polarization and phase structures of the periodically structured light beam. The dependence of period is demonstrated on the length of crystal and the focal length. It is shown that the polarization of the input beam can be used to control the polarization and phase structures of the output beam.
Performing data mining on large waveform datasets of a high-power laser facility is an important way to achieve precise regulation of a device. However, there are currently issues with missing values, noise, and inconsistency in this database of measuring pulse waveform in a current high-power laser facility. In this paper, a UNet of a series residual module is presented to predict the pulse waveform of a front-end chained system in a current high-power laser facility. The designed network is trained on grouped sequences formed by experimentally measuring pulse waveforms of a high-power laser facility. The strategies of relay output and relay loss are employed in training in order to enable the network to predict two kinds of pulse waveforms simultaneously. The trained network achieved an RMSE of 3.38% on the testing set of measuring pulse waveform at a frequency of 1 Hz, and an RMSE of 0.84% on the testing set of setting the voltage of the Arbitrary Waveform Generator (AWG). These results indicate that this method can accurately fill in paired missing values in the waveform database of a high-power laser facility. The main advantage of this method is that it can quickly couple operational parameters for prediction, and this method can be applied to predicting laser performance, cleaning one-dimensional sequences, and maintaining a waveform database.
In this paper, we proposed and fabricated a liquid crystal random-phase plate (LCRPP) for laser beam phase smoothing, while the simulation and experimental verification of LCRPP applied to laser beam’s smoothing were also carried out simultaneously. A new beam smoothing technology named polarisation-phase synchronous smoothing and an improved algorithm based on G-S algorithm specially designed for polarisation-phase synchronous beam smoothing are also proposed based on liquid crystal Pancharatnam-Berry (LC PB) phase devices. Our work on beam smoothing taking advantage of LC PB phase devices not only simplifies the process, deduces the cost and greatly improves the precision and effect of beam smoothing but also provides a novel way for spatial beam shaping, which is of wide applications in laser processing. The simulation parameter setting and experimental model of LCRPP.